Luke

437 posts

Luke banner
Luke

Luke

@Lukelific

Software developer. Building Doclific in public.

Austin, TX Katılım Temmuz 2025
65 Takip Edilen27 Takipçiler
Sabitlenmiş Tweet
Luke
Luke@Lukelific·
If your application consists of a single service, there's virtually no point in building it. Anything that simple could be recreated with AI in a matter of minutes.
English
3
0
6
80
Luke
Luke@Lukelific·
@Gavel_on_X Try powerlifting. It’s more about the numbers. If you have a purpose for each session, it’s much easier than “I’m going to go lift heavy circles until I don’t feel like it anymore”
English
0
0
1
4
Gavel
Gavel@Gavel_on_X·
How do people actually work out consistently? I struggle just to get motivated... Any real tips, none of the “look at your abs” stuff 🤣
English
414
5
341
119.9K
Luke
Luke@Lukelific·
@officer_Lew Ok, but wtf do you expect?! You're running from the Terminator in his prime
English
0
0
0
7
Officer Lew
Officer Lew@officer_Lew·
WATCH🚨: Police officer facing heavy online backlash over his aggressive arrest of this teenager 🚔. Do you see a problem here?
English
18.1K
2.2K
21K
2.2M
Luke
Luke@Lukelific·
@VadimStrizheus lol Manus is still the stupidest acquisition. Like why? Just so they have something to do in a year or two when they decide to kill it?
English
0
0
1
10
Vadim
Vadim@VadimStrizheus·
As a founder, what do you think Anthropic’s next move will be? 1. OpenAI acquired OpenClaw 2. Meta acquired Manus 3. Anthropic??
English
191
4
228
31.5K
Luke
Luke@Lukelific·
@AlexFinn Uh, $200/mo is still cheaper than a $10k+ Mac Studio. At the rate AI, technology, and inflation are progressing, the Mac Studio will be obsolete before it becomes cheaper than subscription.
English
0
0
0
27
Alex Finn
Alex Finn@AlexFinn·
Apple will win the AI war in June 2026 without investing a single penny into AI The war won without a sword They will release the M5 Ultra with 1TB of memory It will be able to run super intelligence more powerful then Opus locally on your desk for free It will fundamentally change the way the entire world operates. EVERYONE with a personalized super intelligent AI employee working on their desk 24/7 $200 a month AI plans will feel antiquated While you've been crapping on Apple for not entering an AI race against 100 other companies shredding trillions of dollars, they ran their own race Personalized local super intelligence will be stage 1 of the singularity. I sincerely hope you're preparing for this
English
715
456
6.8K
937.9K
Luke
Luke@Lukelific·
@garrytan Garry the closet conservative
English
0
0
0
2
Luke
Luke@Lukelific·
I want so badly to use Grok 4.1 fast reasoning instead of Opus 4.6 or GPT 5.3 high, but it's just not feasible for anything really complex at this point. I genuinely enjoy using XAI, but Grok 5 better come soon.
English
0
0
2
49
PJ.dev
PJ.dev@Piyush717177·
I just had an idea. Instead of learning every Linux command, what if we build a layer on top of the kernel not another OS, but an AI layer? You simply describe what you want to do, and the AI writes the shell script for you and executes it right in front of you. That way, you do not need to memorize shell scripting, yet you can still use the command line effectively. It would feel like talking to the terminal instead of typing commands.
English
8
0
12
253
Luke
Luke@Lukelific·
If @netflix was smart, they’d make the next season of Night Agent about the current state of the government. An action show about a hero fighting corrupt pedos in the government would be massively successful and insanely entertaining.
English
0
0
1
27
Luke
Luke@Lukelific·
@DavidOndrej1 Didn’t we just find out Kimi 2.5 was just optimizing for benchmarks, not real world performance? I would have to assume something similar for DeepSeek, but we’ll see.
English
1
0
0
2K
David Ondrej
David Ondrej@DavidOndrej1·
if these DeepSeek numbers are real, the world will never be the same again
David Ondrej tweet media
English
71
37
617
64.1K
Luke
Luke@Lukelific·
@zekramu Better yet, be a man: hold it in until you have a heart attack.
English
0
0
2
11
Luke
Luke@Lukelific·
@kanavtwt Is it a fumble? Clawdbot was cool, but was it really revolutionary? Everything in it had already been done prior, just maybe not all in one place.
English
0
0
3
371
Luke
Luke@Lukelific·
@thematrixb0t @grok explain the thread and give additional context if any
English
1
0
0
2K
matrixbot
matrixbot@thematrixb0t·
Epstein Files: Two children were strangled, died and were buried during their sexual abuse They need to pay for their crimes
matrixbot tweet media
English
557
8.2K
20K
1.4M
Luke
Luke@Lukelific·
@worldofscentss @initjean Buddy your brain doesn’t have an infinite context window. Agents can learn and adapt as well…
English
1
0
0
30
Mr
Mr@worldofscentss·
@Lukelific @initjean Isnt it awesome, your brain can adapt, learn and have infinite context window!
English
1
0
0
16
Luke
Luke@Lukelific·
@worldofscentss @initjean So, essentially, you work within the context of the world? (similar to an LLM) The main difference being a constant stream of consciousness
English
1
0
0
14
Mr
Mr@worldofscentss·
@Lukelific @initjean I’m learning, adapting and creating new ideas. Did you study before how LLMs works ? It comes from deep learning models. You can even ask to ChatGPT actually ironically, it answers well.
English
1
0
0
15
Luke
Luke@Lukelific·
@semsimsek01 @initjean Well, I mean, I literally am senior level and work for a profitable startup in a major tech hub... Funny how you resort to degradation of my qualifications over questioning your own presuppositions.
English
1
0
0
27
Mr
Mr@worldofscentss·
@Lukelific @initjean Yes but you mentioned "cognitive" abilities, but LLMs doesn't do anything cognitive ? It works in the context.
English
1
0
0
17
Luke
Luke@Lukelific·
@v_lakshmi80732 @TheMG3D Welp, if your prompts are anything like your English, I’m not exactly surprised you’re not getting great results
English
0
0
0
23
lakshmi kanth n v
lakshmi kanth n v@v_lakshmi80732·
@Lukelific @TheMG3D What project todo app you cannot build an app using with has novelty even for that matter I’ve used 4.5 opus agents to build a smpl kafka streams which introduced a bug which is unable to figure out for very long time if ai surprised you there might be smting wrg with you skills
English
1
0
0
49
Michael
Michael@TheMG3D·
“we're a mere 12-18 months away from AI being able to do the job of practically anybody sitting at a desk.” Can’t wait for them to layoff a bunch of workers and then scramble to rehire them all after it backfires
Michael tweet media
English
43
38
462
10.9K
Luke
Luke@Lukelific·
@tokiorys @initjean lol I work at a well-established SaaS company with several Fortune 500 clients… if you don’t think it’s capable, it’s a prompt quality and structure problem
English
0
0
0
38
tokiory
tokiory@tokiorys·
@Lukelific @initjean > multimillion line repo bro is cooking world-wide bank system or something.. You already should refactor, any bug will be fatal in this mess > not using it to the fullest I've used Opus4.6 with skills, commands, oh-my-opencode and no, AI is not capable to maintain something big
English
1
0
1
72